On 11/29/2011 01:22 PM, Jim Meyering wrote: > Pádraig Brady wrote: >> On 11/25/2011 10:59 AM, Pádraig Brady wrote: >>> On 11/25/2011 10:13 AM, Pádraig Brady wrote: >>> >>>> Hmm, one of your points above got me thinking. >>>> Might fallocate() fail to allocate an extent with ENOSPC, >>>> but there could be fragmented space available to write()? >>>> That would scupper benefit (2) above :( >>>> I'll ask [email protected] >>> >>> And the response from there is that fallocate() will >>> check the free blocks first, and then try to allocate >>> as a contiguous extent, but that part is not guaranteed. >>> So we still get benefit (2). >> >> However that thread has continued, and it >> was mentioned that benefit (1) might actually >> be detrimental, at least for small files. >> >> http://thread.gmane.org/gmane.linux.file-systems/59092 > > Sigh, indeed. > Do you think it's even worth an option, now?
Having an option to enable is much less interesting. One might auto-enable it for files > 16MB or something. I'm reserving judgement until I understand the issues better. cheers, Pádraig.
